1.1 Sodium-Glucose Cotransporter-2 Inhibitors
Diabetes is a metabolic disease characterised by hyperglycaemia and caused by impaired insulin secretion from pancreatic β-cells and/or insulin resistance in insulin-sensitive tissues, with 90% of diabetic patients having type 2 Diabetes. Despite the development of many anti-diabetic drugs targeting different pathways, treating type 2 diabetes has not achieved therapeutic goals yet. Giving the increasing prevalence of obesity, the prevalence of Diabetes is dramatically increasing which is expected to rise to 366 million worldwide by 2030 (Wild et al., 2004). Sodium-glucose co-transporter (SGLT-2) inhibitors is a novel group of anti-diabetic drugs that work independently of insulin secretion and action by inhibiting glucose reabsorption from the kidney (Chao, 2014). The kidney plays an important role in maintaining glucose homoeostasis by the process of glucose reabsorption that occurs in the proximal convoluted tubule (PCT) through SGLT isoforms. SGLT2 isoform is responsible for 90% of glucose reabsorption and SGLT1 reabsorbs the remaining 10% (Wright, Loo and Hirayama, 2011). Phlorizin is the first discovered SGLT inhibitors that block both of SGLT1 and SGLT2 and proven to improve basal and post-prandial glucose level in diabetic rats (Rossetti et al., 1987). However, phlorizin was not used in humans due to its poor bioavailability and its effect on SGLT-1 which associated with gastrointestinal side effects (Mudaliar et al., 2012).
Canagliflozin then has been approved in the United States in March 2013, and dapagliflozin has been approved in Europe (Chao, 2014). SGLT-2 inhibitors including dapagliflozin, canagliflozin, and empagliflozin act inhibiting up to 30 to 50% of renal glucose reabsorption, therefore lowering the renal threshold for glucose excretion (Mather and Pollock, 2011). The presence of other SGLTs other than SGLT-2 in glucose reabsorption is one of the hypothesis that have suggested to explain the limited effect of SGLT-2 inhibitors but the reason of this in still unclear. Another hypothesis suggested that the saturation of the active secretion of SGLT-2 inhibitors into the proximal convoluted tubules (PCT) is the reason for the limited ability of the drug to inhibit reabsorption at a higher dose (Liu, Lee and DeFronzo, 2012). The novel mechanism of action of the SGLT-2 inhibitor makes it ideal to use as monotherapy and combined therapy with other anti-diabetic drugs without increasing the risk of hypoglycemia due to its effect that is independent if insulin (List and Whaley, 2011). In addition to its effect on decreasing weight due to urine glucose excretion (UGE) which proven to have a beneficial effect on lowering cardiovascular events (Foote, Perkovic and Neal, 2012). However, long-term safety concerns associated with progressive stage of the disease linked the decline in glomerular filtration rate (GFR) and the development of renal impairment with SGLT-2 inhibitors. Other safety concerns related to its effect on bone density, cancer and urinary tract infections (UTIs) (Burki, 2012).
In one double-blind phase three randomised control trial (RCT), canagliflozin was compared with the long-acting sulphonylurea (glimepiride) in patients on metformin but poorly controlled. In addition, other RCT in which dapagliflozin was also compared with a sulphonylurea drug (glipizide) and both SGLT-2 inhibitors were better tolerated and showed a favourable effect on improving glycemic control and lowering cardiovascular risk (Nauck et al., 2011). Canagliflozin 100 mg was non-inferior to glimepiride, And superior to glimepiride for lowering HbA1c when the dose was increased to 300 mg.The trial also showed an increase in HDL, decrease in triglycerides but increase in LDL which might be explained by the increase in activity of lipoprotein lipase (Cefalu et al., 2013). However, in the most recent CANaglaflozin cardiovascular assessment study (CANVAS) which involved a total of 10,142 participants with type 2 diabetes and high risk of cardiovascular events from two trials, canagliflozin also showed a potential decrease in cardiovascular risk but double the risk of imputation compared to placebo group (Neal et al., 2017).
Metformin which lowers plasma glucose level by inhibiting gluconeogenesis is still considered as first-line therapy for type 2 diabetes with the best evidence of tolerability, lower side effect, and best effect on lowering macrovascular and microvascular events among other anti-diabetic drugs. Although the exact molecular mechanism of action is not clear, it was reported that metformin enhances cellular insulin-dependent glucose uptake and lower plasma triglyceride alongside with its main action on decreasing hepatic glucose output (Galuska et al., 1994). It has been proven in the literature that metformin inhibits complex 1 of the respiratory chain. Even though it was suggested that metformin has a novel action on activating AMPK in intact cell and in vivo in an effect that is not related to energy depletion that is caused by the drug effect on inhibiting gluconeogenesis (Hawley et al., 2002), more recent data proved the opposite. Metformin failed to activate AMPK in mutant γ subunit that does not bind to AMP/ADP supporting the effect of energy depletion state in metformin AMPK activation (Hawley et al., 2010).
Recent data report that canagliflozin at human plasma concentration used in clinical trials activate AMPK by increasing the cellular AMP and ADP through inhibiting Complex I of the respiratory chain. But such effect was not reported with dapagliflozin or dapagliflozin. Such effect might add to the beneficial therapeutic effect of canagliflozin that is unrelated to SGLT-2 action on lower plasma glucose via kidney excretion (Hawley et al., 2016).
1.2 AMPK
1.2.1AMPK structure and regulation
5’ Adenosine monophosphate-activated protein kinase (AMPK) is an enzyme consisting of three subunits (α, β, and γ) with each subunit playing a role in the activity and stability of the protein complex (Hardie et al., 2012). AMPK- α is the catalytic subunit in which the activation of AMPK occurs via the phosphorylation of the threonine residue in the kinase domain (Thr172) at the N-terminus. It is also the most intensively studied subunit among these subunits, while AMPK-β and AMPK-γ are regulatory subunits(Hawley et al., 1996). AMPK β subunits contain glycogen binding site which inhibits AMPK activity while AMPK γ contains 3 binding sites in which one of them AMP molecule bind tightly and two of them being the competitive binding site for AMP, ATP or ADP (Xiao et al., 2007). Different subunit has different isoforms that were suggested to assure highly conserved cellular expression and distribution, which makes AMPK a potential target for selective activation (Salt and Palmer, 2012). Binding of AMP and ADP to γ subunit activate the phosphorylation of Thr172. In contrast, the allosteric competitive binding of ATP (when ATP: ADP ratio increase) antagonise the phosphorylation of Thr172 that was initiated due to energy depletion (increase in ADP: ATP ratio) (Davies et al. 1995). Liver kinase B1 (LKB1) and Ca2+/calmodulin-dependent protein kinase kinase β (CaMKKβ) are to the upstream kinase that mediates the phosphorylation of Thr172 within catalytic α subunit (Hawley et al. 1996).While LKB1 cause activation of AMPK in through alteration of AMP and ADP level, CaMKKβ represent an alternative pathway that activates AMPK in counter to increased cellular Ca+2 level (Hardie et al., 2012).
1.2.2 AMPK effect on Lipid metabolism
AMPK is considered to be the first energy sensor in eukaryotic cells and is activated in response to energy depletion when the ratio of adenosine triphosphate (ATP) to adenosine monophosphate (AMP) is reduced (Hardie et al., 2012). AMPK activation affects many cellular metabolism pathways such as cholesterol, glycogen and fatty acid synthesis to maintain bioenergetic state by normalising cellular ATP level. This effect happens through increasing catabolic pathways which include lipolysis, fatty acid oxidation and inhibiting anabolic pathways such as cholesterol and fatty acid synthesis to increase ATP level (Carling et al.,1987). HMG- CoA reductase (HMGR) which is a rate limiting enzyme for cholesterol synthesis is inhibited and phosphorylated by AMPK (Burg and Espenshade, 2011).
One of the key pathway regarding AMPK effect on lipid metabolism is its phosphorylation and inhibition of Acetyl-CoA carboxylase (ACC), which is the rate-limiting enzyme for fatty acid oxidation. ACC act by converting acetyl CoA to malonyl CoA which promotes fatty acid synthesis and prevent fatty acid oxidisation by two different mechanisms. Two isoforms of ACC are expressed in human, ACC 1 which is mainly found in liver and adipose tissue and ACC 2 which is believed to be responsible for fatty acid oxidation in cardiac and skeletal muscle (Merrill et al. 1997, Tong, 2005). Malonyl CoA is considered as the building block for fatty acid synthesis by providing acetyl group (Tong, 2005). In addition to its allosteric inhibition of carnitine palmitoyl transferase 1 (CPT1), which transfer fatty acid to mitochondria for oxidation (McGarry 1995). Sara ref Therefore AMPK inhibitory effect on ACC prevent fatty acid synthesis and release the inhibitory effect on CPT1 to promote fatty acid oxidation to increase ATP level. Additionally, it was reported that AMPK activates malonyl CoA decarboxylase which facilitates the conversion of malonyl CoA to acetyl CoA (Assifi et al. 2005).
Different catalytic isoforms of AMPK were reported to be expressed dominantly in different cell type. While AMPK α1 was found to be the mainly expressed in cultured 3T3-L1 adipocytes (Salt, Connell, and Gould 2000), AMPK β1 was reported to be mainly expressed in human endothelial cells (Wang, Zhang, et al. 2010). Caloric restriction was found to activate AMPK in adipocytes (Daval et al. 2005) and its activation was found to inhibit adipogenesis by decreasing the expression of C/EBPα and PPARγ which are required for the initial adipogenic transcriptional cascade (Giri et al. 2006). However, the exact mechanism of AMPK effect on inhibiting adipogenesis is still unclear.
1.2.3 AMPK effect on Glucose transport and carbohydrate metabolism
The regulatory effect of AMPK in adipocyte, hepatocyte, myocytes, neurones and pancreatic β cells has been demonstrated to involve several pathways in various tissues types. Such effect occurs either via fast response which involves mainly phosphorylation of various proteins and more delayed or long term effect via gene expression (Hardie et al., 2012). One of these pathways is glycogen synthesis in which AMPK activation inhibits and phosphorylate glycogen synthase enzyme in the liver and skeletal muscle (Bultot et al., 2012).
Glucose uptake is mediated by the translocation of glucose transporter 4 (GLUT4) which facilitate glucose influx in the cell through a vesicular fusion of vesicle contain GLUT4 with plasma membrane which carries glucose inside the cells (Yamaguchi, 2005). Signalling pathway causing the translocation of GLUT 4 beside the identification of the involvement of p85/p110 phosphatidylinositol 3-kinase (PI3K) activation have not been discovered yet despite the extensive research have been done (Heller-Harrison et al., 1996). While AMPK activation demonstrated to increase basal and insulin stimulated glucose uptake In skeletal muscle, it showed an inhibitory effect on glucose transport in adipose tissue. (Ojuka, Nolte, and Holloszy 2000).
5’-aminoimidazole-4-carboxamide ribonucleoside (AICAR), an adenosine monophosphate (AMP) analogue that enters the cell through Adenosine transporter and gets phosphorylated to ZMP by adenosine kinase, was reported to increase basal but inhibit insulin-stimulated glucose uptake in adipocytes (Salt, Connell and Gould, 2000). However other evidence showed that AICAR inhibits both basal and insulin stimulated (Gaidhu et al., 2011). Overall, AMPK activation in adipocytes showed an inhibitory effect on insulin- stimulated glucose uptake. However molecular mechanisms are not fully understood.
1.2.4 AMPK role in type 2 diabetes and obesity
Activation of AMPK by anti-diabetic drugs such as biguanide (metformin) and thiazolidinedione (rosiglitazone) has been demonstrated to reverse the defects caused by metabolic syndrome which involve obesity, type 2 diabetes and hypertension (Scott et al., 2008). The same effect of AMPK activation was seen in restricted caloric intake which emphasises the association between obesity and type 2 diabetes (Viollet and Andreelli, 2011). As obesity increase adipose tissue by hypertrophy and hyperplasia which refers to increase in size and number of adipocyte respectively, AMPK activation was reported to have an exclusive inhibitory effect on adipocyte hypertrophy. Such observation was reported by comparing mice deficient in the α2 catalytic subunit with the wild type (Villena et al. 2004). AMPK activation also reported with statin which showed a protective effect on cardiovascular system (Sun et al., 2006). Therefore, AMPK has been studied for a potential therapeutic target for type 2 diabetes and other metabolic diseases (Bijland et al., 2013).
1.3 Aim
The aim of this study is to investigate the effect of sodium-glucose cotransporter-2 inhibitors (canagliflozin, dapagliflozin and empagliflozin) on AMPK activation in the 3T3-L1 cultured adipocyte. In this study, we incubated 3T3-L1 adipocyte with SGLT-2 inhibitors and AICAR for a variable duration with and without insulin to investigate the effect on AMPK cascade and glucose transport.